Are there differences in game profiles within the same overall position ?
FIXTIVE STORY :
We are looking for the best player available for our team. We need a central position player with main offensive qualities. Defensive qualities are secondary, as we already have one of the best defences in the league.
Clusters ?
It is necessary to express the playing time data in order to make a relevant comparison.
Activité Offencive : cumul of disposals season divided by minutes played.
Activité Defencive : cumul gains (defections + interceps) divided by minutes played.
This graph classifies the different athletes into three distinct clusters using the Kmeans method. The data appears to be grouped as “mixed activity”, “attack dominated activity”, “defense dominated activity”
profitability of the player in attack ?
Profitability is the balance between the amount of work the player does and his efficiency. Having a player with a lot of game action but little efficiency is not interesting for us. Similarly, if a player is very efficient but does not do much in the game, then he does not affect the dynamics of the game.
TOP 3: Price, Nankivell and Jovic, are the best performing players in the Kmean category “Mixed activity”,
TOP 3: Garbin, Fowler and Wallace, are the best performing players in the Kmean category “Attacking dominance activity”,
TOP 3: Mannix, Delaney and Sterling, are the best performing players in the Kmean category “Defending dominance activity”.
Finally, for our team, the most relevant choices seem : Garbin, Fowler and Wallace!
And about defense ?
This graph shows the profitability of the top 3 players isolated as mentioned above. This allows us to determine which player has done the most work efficiently balance. In this case, we focused on the defensive activity.
📢 WE WANT GARBIN,S !!!!
This player fits our needs best. In addition to his offensive activity of 62 actions per game with an efficiency of 95%. He also knows how to be efficient in a defensive system with 3 actions per game, 93% efficiency 💪🏽.